Soft modular robots

Modular robots are capable of adapting their morphology to various tasks and environments, which makes them more versatile, flexible and robust compared to fixed-bodied ones. However, most of the current modular systems have hard building blocks and highly rigid connection mechanisms, which leads them to lose essential flexibility as the assembly grows in size. Therefore, modular designs’ improved controllability and stability comes at a cost of flexibility. In order to improve the structure’s adaptation to environmental changes, softness on the module level might be beneficial. What if modules were to become completely soft and the robot could benefit from almost infinite deformability? 

The goal of this project is to look at how mechanical module softness can increase the efficiency and the capabilities of a modular robot. However, coping with softness requires fundamental rethinking in the way the modules are built. In this study we will investigate soft modules that feature a reversible connection mechanism, active deformation and sensing.  We will also design and develop novel soft technologies, smart sensing and actuation.

Soft connection mechanism

We developed a connection mechanism for soft modules based on electroadhesion that improves connection efficiency in a wide range of module softness (e.g. from 30 to 130kPa-1), confers high tolerance to misalignment (e.g. the attractive force decreases less than 40% for 70% misalignment) and allows for easy detachment. This mechanism does not reduce the softness of the module membrane, is suitable for light-weight modules (<10g) and could be useful for stochastic modular robotic systems. Stochastic robotic systems are promising in reducing module size and scaling systems to large numbers.

(Left) Module mockup featuring electroadhesion, (right) two modules connected.

Publications

Soft Cells for Programmable Self-Assembly of Robotic Modules

J. M. Germann; A. Maesani; R. Pericet Camara; D. Floreano 

Soft Robotics. 2014. Vol. 1, num. 4, p. 239-245. DOI : 10.1089/soro.2014.0005.

Stretchable Electroadhesion for Soft Robots

J. M. Germann; B. E. Schubert; D. Floreano 

2014. IEEE/RSJ International Conference on Intelligent Robots and Systems, Chicago, Illinois, USA, September 14-18, 2014. p. 3933-3938. DOI : 10.1109/IROS.2014.6943115.

Programmable Self-assembly with Chained Soft Cells: An Algorithm to Fold into 2-D Shapes

J. Germann; J. Auerbach; D. Floreano 

2014. Simulation of Adaptive Behavior 2014, Castellón, Spain, July 22-26, 2014. p. 220-229. DOI : 10.1007/978-3-319-08864-8_21.

Soft Cell Simulator: A tool to study Soft Multi-Cellular Robots

J. M. Germann; A. Maesani; M. Stöckli; D. Floreano 

2013. IEEE International Conference on Robotics and Biomimetics, Shenzhen, China, December 12-14, 2013. p. 1300-1305. DOI : 10.1109/ROBIO.2013.6739644.

An Active Connection Mechanism for Soft Modular Robots

J. M. Germann; M. Dommer; R. Pericet Camara; D. Floreano 

Advanced Robotics. 2012. Vol. 26, num. 7, p. 785-798. DOI : 10.1163/156855312X626325.